Annals of the Society for
the History of Economic Thought, Vol.36 (Oct. 1998).
Recent Japanese Studies in the Development
of Keynes's Thought: An Evaluation
Toshiaki HIRAI
The object of this paper is to evaluate, from
the point of view of my own stance on the issue, some recent studies which have
appeared in Japan
Clarification of Keynes's theoretical
development is, as we know, an exceedingly difficulttask. To begin with, we
have to contend with the existence of three Keynesian schools, each at odds
with the other two, notwithstanding that all base their approaches on the General Theory. These are Income-Expenditure Approach
Keynesians, the Disequilibrium Approach Keynesians and the Post-Keynesians. The
controversies between these schools have come about not only because their
interpretations of the General Theory
differ so greatly, but also because their views on how the theory should be
revised are remarkably divergent. Inevitably, this circumstance must have no
small effect on how researchers who attempt to clarify the development of
Keynes's ideas understand the General Theory
itself (the end-point of the process under consideration). Secondly, there also
exist several rival understandings of Keynes's Treatise
(the starting-point). Thirdly, as a natural consequence of the first two facts,
we have a number of competing views of the relation between the theories of the
two great books. Any resercher who attempts to delineate the evolution of
Keynes's theorising will inevitably be influenced, consciuosly or
unconsciously, by the reverations of these controversies.
1. A Treatise on Money and
the General Theory
A. The Treatise
I understand the central theoretical structure
put forward in the Treatise ― let
us call this 'Keynes's own theory' ― as follows (as will be explained later,
the Treatise also contains a second theory,
which may be referred to as the 'Wicksellian theory'):
The structure can be described as a dynamic
process composed of 'Mechanism 1' (the determination of the price level of
consumer goods in each period - this is in substance the same as 'the first
fundamental equation'), 'Mechanism 2' (the determination of the price level of
investment goods in each period - this is determined either in the stock market
or as the demand price of capital goods, the former case being related to the
'bearishness function' theory), and the 'TM supply function' (which describes
the behaviour of entrepreneurs as being such that, if they make a profit (loss)
in the current period, they expand (contract) output in the next).
The distinctive feature of this interpretation
lies in its seeing the Treatise as
articulating a dynamic process which includes the determination of both the
price levels and the volumes of output. We stand by this interpretaion because
sufficient documentation to demonstrate its validity exists. As will be seen
later, the development of Keynes's thinking after the Treatise
cannot be understood unless we accept this interpretation. Akashi
The orthodox interpretation of the Treatise theory, by contrast, understands it as expressing
'a theory of the determination of the price levels under a fixed volume of
output (and this fixed at the level of full employment) and as lacking a theory
of output'. (Even when the TM supply function is referred to, it is treated as
unimportant.) This interpretation, which is accepted by both the Income.
Expenditurists and the Disequilibriumists, is also widely accepted by other
reserchers, Hishiyama (1933) pays attention, so far as the treatment of the
commondity markets in the Treatise is
concerned, to the determination of the price levels of consumption and
investment goods, but neglects Keynes's arguments concerning the volume of
output. (As discussed below, Professor Hishiyama grasps Keynes's development in
terms of the relation between the money and the natural rates of interest.
Hishiyama (1965), however, explicitly deals with the changes in the volume of
output through the changes in prices emerging as a result of the occurrence of
profits.) The same stance is to be seen in Kanoh (1992), Kojima (1997) regards
the Treatise theory as dealing with
relations within periods, and on the whole does not accept the idea that it
possesses a dynamic attribute in the form of a sequential analysis (though he
does touch upon the TM supply function). Asano (1987) adopts a delicate
position, maintaining that the main Treatise theory
is concerned with an account of the determination of the price levels, but also
recognising that the Treatise
contains in addition a theory dealing with changes in the volumues of output.
However, he insists that the two theories are logically inconsistent, each
belonging to a completely separate theoretical system, whereas I regard the two
theories as being coherently connected, and as constituting a dynamic model
based on sequential analysis.
B. The General Theory
As I understand it, the essence of the General Theory can be summed up as 'the monetary economics
of underemployment equilibrium'. Here 'monetary economics' is understtod to
have the following features:(1) it stems from dissatisfaction with the
dichotomy of classical economics ― both the dichotomy in the standard sense and
that recognisable in the classical theory of the rate of interest; (2) it
arises also through criticism of the 'Wicksell Connection', to be discussed in
Section 2; (3) it analyses an economy in which money plays a vital role, as is
shown by, for example, presentation of the theory of liquidity preference, and
inquiry into the nature and significance of money by means of the concept of
'ownrates of interest' , 'Underemployment equilibrium' is intended to convey
the following ideas: (i) a state of the economy below full employment; (ii) a
kind of 'equilibrium'; (iii) a kind of 'stability'; (iv) allowance for
fluctuations within a certain range.
Keynes puts forward his 'monetary economics of
underemployment equilibrium' in the form of a system of simultaneous equations,
while also considering the causal relations between variables. It should be
noted that in the General Theory Keynes
characterises the market economy as possessing two aontrasting attributes -
stability, certainty and simplicity on the one hand, and instability,
uncertainty and complexity on the other. In my view, any understanding of the General Theory which ingores either attribute must be
inevitably inaccurate.
The most important theory which Keynes puts
forward in the General Theory is. of course, the
model showing how the level of employment is determined. This is based on three
'untimate independent variables'; (1) the wage unit; (2) the quatity of money;
(3) the three fundamental psychological factors, these being the propensity to
consume, liquidity preference, and the schedule of the marginal efficiency of
capital. Keynes argues that the level of employment is determined where the
aggregate demand function (derived from these variables) meets the aggregate
supply function (derived from the 'first postulate of the classical economics')
In this regard the following four points
should be particularly noted. First, Keynes is sure that the 'stickiness' or '
rigidity' of money wages contibutes to the stability of the economic system.
Second, Keynes unambiguously rejects the concepts of 'general price-level' and
'national devidend', declaring that two fundamental units of quantity only ―
quantities of money value and quantities of employment ― should be used in
developing the theory of employment. Third, Keynes regards prices as flexible
(his opinion being that they depend partly on the wageunit and partly on the
volume of employment). Fourth, Keynes argues that the working of the market
economy depends on a variety of psychological and expectational factors which
affect people's attitudes to the future; short term expectations, long term
expectations, liquidity preference due to the speculative motive, and user
cost. It is with these points in mind that I refer to the theoretical framework
of the General Theory as the 'Heterogeneity -
Expectations Approach' and attempt to reconstruct it as a model equipped with a
two-tier adjustment mechanism (micro and macro structures). Leaving to one side
any questions concerning the appropriateness of my reconstruction. I am sure
that the heterogeneity expectations viewpoint sheds much useful light on the
theoretical structure actually developed in the General
Theory (see Hirai, 1981).
Suprisingly, few researechers studying the
formation of Keynes's economics present a systematic account of the theoretical
structure of the General Theory as distinct from
inquiring into the meaning of certain individual chapters, such as Chapter 17.
This may be due either to the fact that disagreement as to understanding of
what is put forward in the General Theory
― one of the root causes of the conflicts between the above mentioned Keynesian
schools ― makes it difficult for studies of Keynes's development to exmine the
theoretical structure directly, or else to the assumption that the meaning of
the General Theory has already been fully
elucidate. Thus recent studies such as, for example, Okada (1997), Kojima
(1997) and Yoshida (1997), contain little or no systematic analysis of the General Theory. Rather, studies of Keynes's ideas generally
focus their attention on the Treatise, on
the theories of Keynes's contemporaries such as Robertson and Hawtrey, and on
Keynes's own earher theories. That is not in itself undesirable, yet to spell
out his or her own understanding of the General Theory
remains an indispensable task for any researcher.
C. Keynes's
Theoretical Development
How did Keynes alter and develop his economic
theory between the Treatise and
the General Theory? After presenting my own
view in the form of 'hypothesis', I will exmine other reserchers' views
concerned.
(Hypothesis 1)
Keynes remained in the world of the Treatise until just before the manuscript 'The Parameters of
a Monetary Economy' written at the end of 1932.
Remained in the world of the Treatise means 'stuck to 'Keynes's own theory'. Recall that
a second theory ― the 'Wicksellian theory' ― was to be abandoned by Keynes from
immedeately after the Treatise on. Worthy
of attention concerning Keynes's Worthy of attention concerning Keynes's own
theory is the fact that Keynes held on to the TM-supply function in spite of
critisism from, among others, the 'Cambridge Circus'. In this respect, Yoshida
(1997) is in accord with my argument. (The difference between Professor Yoshida
and I lies in his playing down the General Theory,
with the assertion that it represents no more than a pared-dowm version of the Treatise, while I argue that a great change occurs between
the two.) Virtually in line with my own view of Keynes's development is that of
Akashi (1988), as is indicated in the following statements; 'the transformation
of Keynes's thought process can be found in the shift of his concern from the
fluctuations in output to the “equilibrium” level of output attained as a
result of them (p.106); 'the methodological transformation [is] from a
sequential analysis to an equilibrium analysis'(p.111).
(Hypothesis 2)
The 'qualitative transformation' from the
world of the Treatise to that of the General Theory takes place in the manuscript 'The Pramenters
of a Monetary Economy'
I see support for this hypothesis in the
following features of the manuscript; (1) the adoption of the notion of
'investment saving equilibrium'; (2) the formulation of a supply schedule as a
function of the price complex' (the essential disappearance of the TM-supply
function); (3) the presentation of a system of simultaneous equations equipped
with an ides similar to the multipher theory.
The manuscript adopts an equilibrium analysis,
abandons a sequential analysis, and incorporates important components to be
used later in the General Theory, so that it can
accurately be said to be completely independent of the Treatise.
On the other hand, however, it is still some way short of the General Theory in that it lacks an account of how the level
of employment is determined. It is for these reasons that I indentify 'The
Parameters of a Monetary Economy' as the crucial turning point.
Researchers who emphasise the importance of
this manuscript in Keynes's development process include Hishiyama (1993),
Kojima (1997) and Okada (1997). Professor Hishiyama uses 'the Wicksellian mode
of adjustment to mean the adjustment of the money rate of interest to the
natural rate, and regards the Treatise as
exemplifying this mode of adjustment. He maintains that Keynes's thinking
underwent a great transformation between the Treatise
and the General Theory as a result of his coming
to accept Sraffa's idea that the natural rate of interest adjusts to the money
rate of interest, and that 'The Parameters of a Monetary Economy provides
evidence for this. ( I agree with his view that the Treatise
has the Wicksellian mode of adjustment, but I would stress again that it also
contains another theory ― 'Keynes's own theory' ― and that Keynes drops the
Wicksellian mode of adjustment
(the
'Wicksellian theory') immediately after the Treatise.)
Kojima (1997) attempts to confirm Professor Hishiyama's hypothesis by
examinining Sraffa's 1932 review of Hayek's Prices and Production, but his
reasoning is open to several questions. In the first place, it rests more on
conjecture than on documentation. Secondly, Professor Kojima identifies the
'natural rate of interest' with Sraffa's 'commondity rate of interest' defined
in terms of the price relations between the spot and the forward markets
(p.10), and then treats it as if it were identical to Wicksell's 'natural rate
of interest'. Thirdly, although he insists that 'The Parameters of a Monetary Economy'
emerged as a result of Sraffa's critique of Hayek's theory, the assertions he
makes in support of this contention ― regarding (i) the adoption of the concept
of a complex', and (ii) the exogeneity of the money rate of interest ― are
unpersuasive. In the manuscript, the money rate of interest is in fact treated
as an endogenous variable, as is evident in the formulation 'p = A (・)' where p is the (money) rate of
interest, A (・) the state of liquidity preference, and M the
quantity of money, which is an exogenous variable Okada (1997) similarly
emphasises the significance of 'The Parameters of a Monetary Economy', but from
a quite different angle, asserting that 'we can confirm that Keynes, in the
late 1932 on the way from the Treatise to the
General Theory, accomplished the most
radical change, when the Keynesian. Revolution essentially took place'.(p.204)
I agree with the first part of this statement, but do not accept the second
part on the following grounds. (1) the manuscript sets out no investment theory
leading up to the investiment theory of the General Theory;
(2) the manuscript offers no accout of how the level of employment is
determined. In addition, Professor Okada overlooks the fact that Keynes's
methodology has already shifted from sequential analysis to equlibrium
analysis.
Although he detects certain progressive
elements in 'The Paramenters of a Monetary Economy' which points towards the General Theory, Asano (1987) nonetheless considers that
Keynes is here still 'adhering to the theoretical framework of the Treatise' (p.89). For, according to Professor Asano, at this
stage Keynes still clings to the TM supply function, continues to use prices as
the principal variables of the system, and adopts the liquidity preference
theory which, lacking a speculative motive for holding money, represents the
remnants of the bearishness function approach. Here I would suggest that
Prefessor Asano fails to pay sufficient attention to the fact that in the model
of this manuscript the TM supply function does not play any substantial role,
and that the method of analysis changes completely. In general, Professor Asano
evaluates the manuscript from a perspective which looks back to the Treatise, whereas I evaluate it from a forward-looking
stance. At the same time, though, he does acknowledge that 'there was a
decisively important progress in thought in the late 1932' (p.100), as
witnessed, in his opinion, by the article 'The Monetary Theory of Production',
which appeared as Keynes's contribution to Spiethoff Festschrift. In
maintaining this positon Prefessor Asano is concerned to emphasise, above all,
the 'negation of the neutrality of money'. Concerning this I would point out
that this matter was discussed in the first lecture of the Michaelmas term of
1932, and that the core of that term's lecture course was an exposition of the
same model as is put forward in 'The Parameters of a Monetary Economy'.
(Hypothesis 3)
In the three manuscripts of 1933 we encounter
the origins of Chapter 3, 'The Principle of Effective Demand', of the General Theory, as they discuss the equilibrium and
stability conditions of the level of employment. Yet we also detect several
logical ambiguities, so that it is clear that Keynes is still engaged in
'searching efforts'.
The three manuscripts of 1933 have been
examined in detail in both Asano (1987) and Hirai (1987). As Prefessor Asano's
exegesis is close to the mark and there is no great divergence between our
interpretations, I would here like just to proceed with my own view. I regard
the First Manuscript as a system of simultaneous equations which determines the
level of employment, in consequence of which I evaluate it in relation to
Chapter 3 of the General Theory, Akashi
As to the Second and Third Manuscripts, I see
as their essential feature the argument for the stability of the equilibrium
position. Asano (1987) is accurate when in his evaluation of the Second
Manuscript he points out that 'Although he continued to take pains to give
readers the impression of the continuity with the Treatise,
Keynes in fact took a rather decisive step away from it' (p.130).
It is around the period of the Second
Manuscript that Keynes adopts the 'first postulate of the classical economics',
Asano (1987) maintains that this implies a decisive shift in Keynes's thinking
― 'a great transformation of the supply function from that in the Treatise to a new idea based on marginal analysis' (p.175) ―
and that this transformation is attributable to Kahn's persuasion. I agree with
Professor Asano's analysis. Yoshida (1997), on the other hand, insists that the
first postulate is no more than 'noise' which was never incorporated into the
foundations of Keynes's theory, and furthermore that the Cambridge Circus
obstructed the natural development [of the Keynesian Revolution]' (p.183). I do
not concur with this view. Professor Yoshida fundamentally believes the
concepts of 'demand, supply and equilibrium' to be harmful to sound economic analysis.
It is from this point of view that he looks at Keynes's development. He holds
the Treatise in highest eteem, for he finds
it the only book free of such concepts. When we turn to the General Theory, however, we see that Keynes makes much use
of these concepts. (As noted earlier, Keynes's method also shifts, from
sequential to equilibrium analysis)
That is why the Treatise
receives so much attention from Professor Yoshida, at the expense of the General Theory, which he rather neglects. This stance ties
in closely with his rejection of the first postulate, Professor Yoshida
repeatedly emphasises the importance of 'the original text', yet he
nevertheless seems to ignore it.
With regard to the period following the end of
1933, I would like to draw attention to the following points. (1) It was from
late 1933 to early 1934 that Keynes essentially completed the consumption
theory and substatially imprpved the investment theory. (2) In Hirai (1987,
Chapter 8-10, 1997, Chapters 12-14). I examined various manuscripts from 'The General Theory' of spring 1934 to the galleys of the General Theory, I think it important that these should be
examined in relation to the theoretical structure put forward in the General Theory - which makes it incumbent on reserchers to
spell out their own understandings of the General Theory
in advance. In this light, it is my impression that the manuscripts of this
period have up to now received very little in the way of really serious
examination.
2. The 'Wicksell
Connection'
In parallel with studies of the development of
Keynes's economic theory, several studies of the economic theories of his
contemporaries have emerged in recent years. In this section I would like to
discuss some studies relating to one contemporary stand of economic thought ―
the 'Wicksell Connection', I use this term to denote either the group of
economists who accepted (critically) and developed Wicksell's theory of
cumulative process in one way or another (Lindahl, Myrdal, Mises, Hayek,
Robertson, Keynes of the Treatise,
Hawtrey, and others) or the strand of ecomic theories generated by these
economists. This movement rejected 'the quantity theory of money, Say's Law,
and the cllasical dichotomy' and attempted to construct a monetary economics.
The earliest of the receny studies the Wicksell Connection to be considered
here is Akashi Stockholm
School (Lindahl and Myrdal), and on
the other the Cambridge
School Staffa criticised Wicksell Keynes indirectly through his
criticism of Hayek, and that Keynes grasped this and was thus prodded to shift
his thinking in the direction which was eventually to lead to the General Theory.
Kojima (1997) denies the validity of the
Wicksell Connection concept. In its place, he puts forward an alternative
concept called the 'Hawtrey Connection'. His argument, however, it less than
persuasive. In the first place, Professor Kojima himself refers to and
discusses the connection between the Treatise and
Wicksell's theory, while his discussion of the Hawtrey Connection never rises
above conjecture, Secondly, his definition of the Hawtrey Connection (p.100)
seems to be narrow. It is defined in terms of (i) a theory of the essence of
money, and (ii) 'a circular-structural understanding of monetary economy' (thus
Professor Kojima's interpretation of the production structure of the Treatise again lacks the TM supply function). The problem
with this definition is that it relates not to the connection between Hawtrey,
the Treatise and the General
Theory, but only to the connection between Hawtrey and Keynes in the
1920s. (It is the former connection, in my view, that really matters.) Though
Professor Kojima allocates three chapters to Hawtrey, the argument in Chapters
4 and 5 in fact has little to do with the definition which appears in Chapter
3, Yoshida (1997) likewise criticises the con-cept of the Wicksell Connection.
This stems from the fact that, with respect to the Treatise,
Professor Yoshida rather overemphasises 'Keynes's own theory' (in my
terminology), and neglects the Theatise's other theory ― the 'Wicksellian
theory'. This stance, though not exactly wrong, is nevertheless somewhat
biased.
I would now like to refer briefly to an
interpretation of the positions of Wicksell's theory and the Treatise in the history of economic thought due to Okada
(1997). Being concerned with neither the differences between the classical and
neo-classical schools (plutology versus catallacties), nor the theories held in
common by the two schools (the classical dichotomy, Say's Law, and the quantity
theory of money), nor the differences between these two schools and the
Wicksell Connection (real economics versus monetary economics). Professor Okada
regards both Wicksell's theory of cumulative process and the Treatise as belonging to the 'classical school', considered
as accepting the quanitity theory of money and the classical dichotomy. Then he
sees the General Theory as freeing itself of the
'classical school'.
I conclude this paper with a restatement of my
own view of the relations between the Wicksell Connection, the Treatise and the General Theory.
The Treatise contains two distinct theories
― 'Keynes's own theory' and the 'Wicksellian theory'. In the immediate post Treatise period, Keynes came to pay attention exclusively to
the first of these theories. Through intensive reflection upon it, together
with his adoption of an equlibrium analysis approach, he came to formulate the General Theory as the 'monetary economics of underemployment
equilibrium'. Thus the General Theory
represents a monetary economics which differs from that of the Wicksell
Connection. In short, it was not only 'classical economics' but also the
Wicksell Connection that Keynes attacked in the General
Theory.
Notes
1) Fujii (1994)
provides an examination of recent studies in the field, inclusive of foreign
papers.
2) The framework
for the controversy over the Wicksell Connection has already been prepared by
Komine (1997). The controversy is likely to get more heated from now on.
References
Asano, E.
(1987), Keynes Ippan Riron Keiseishi (The Development Process Leading to
Keynes's General Theory), Nihon Hyoronsha.
Fujii, K.
(1994), 'Recent Studies on the Formation of Keynes's Theory,' Annals of the
Society for the History of Economic Thought, Vol.32.
Hirai, T.
(1981), Keynes Ippan Riron no Saikouchiku (A Reconstruction of Keynes's General Theory), Hakutoh Shobou.
Hirai, T.
(1987), Keynes Kenkyu (A Study of Keynes), University of Tokyo Press
Hirai, T.
(1990), 'The Wicksell Connection (I) (II)'. Sophia Economic Review, Vol.36,
No.1 and No. 2.
Hirai, T. (1997,
1998), 'A Study of Keynes's Economics(I)-(IV), sophia Economic Review, Vol.43.
No.1 and No. 2, and Vol.44. No.1 (forthcoming) and No.2 (forthcoming).
Hishiyama, I. (1965), Kindai Keizaigaku no Rekishi (History of
Modern Economics), Yuushindo.
Hishiyama, I.
(1990), Quesnay kara Sraffa e (From Quesnay to Sraffa), University of Nagoya Press
Hishiyama, I.
(1993), Shaffa Keizigaku no Gendaiteki Hyouka (Modern Ecaluation of Shaffian
Economics), University
of Kyoto
Kanoh, M.
(1992), Keynes Kahei Keizairon (Keynes's Monetary Economics), Kourosha.
Matsukawa, S.
(1991), Keynes no Keizaigaku (Keynes's Economics), Chuo Keizaisha.
Okada, M.
(1997), Kyoshiteki Keizai RIron no Kiseki (The Development of Macroeconomic
Theory), University
of Nagoya
Kojima, H.
(1997), Keynes Riron no Gensen (The Origins of Keynes's Theory), Yuuhikaku.
Komine, A.
(1997), 'Survey of a Monetary Economics ― the Case of the Wicksell Connection.'
Annals of the Society fo the History of Economic Thought, Vol.35.
Yoshida, M.
(1997), Keynes, Nihon Keizai Hyouronsha.
